OBJECTIVE The present randomized controlled study investigated whether sonographic needle guidance affected the outcomes of intra-articular injection for osteoarthritis of the knee. METHODS Ninety-four noneffusive knees with osteoarthritis were randomized to injection by conventional palpation-guided anatomic landmark injection or sonographic image-guided injection enhanced with a 1-handed mechanical (the reciprocating procedure device) syringe. After intra-articular placement and synovial space dilation were confirmed by sonography, a syringe exchange was performed, and 80 mg of triamcinolone acetonide was injected with the second syringe through the indwelling intra-articular needle. Baseline pain, procedural pain, pain at outcome (2 weeks and 6 months), responders, therapeutic duration, reinjection rates, total cost, and cost per responder were determined. RESULTS Relative to conventional palpation-guided anatomic landmark methods, sonographic guidance for injection of the knee resulted in 48% reduction in procedural pain (P < 0.001), a 42% reduction in pain scores at outcome (P < 0.03), 107% increase in the responder rate (P < 0.001), 52% reduction in the nonresponder rate (P < 0.001), a 36% increase in therapeutic duration (P = 0.01), a 13% reduction ($17) in cost per patient per year, and a 58% ($224) reduction in cost per responder per year for a hospital outpatient (P < 0.001). CONCLUSIONS Sonographic needle guidance reduced procedural pain and improved the clinical outcomes and cost-effectiveness of intra-articular injections of the osteoarthritic knee.